Search

Background Correct certification of cause of death by physicians (i.e. completing the medical certificate of cause of death or MCCOD) and correct coding according to International Classification of Diseases (ICD) rules are essential to produce quality mortality statistics to inform health policy. Despite clear guidelines, errors in medical certification are common. This study objectively measures the impact of different medical certification errors upon the selection of the underlying cause of death. Methods A sample of 1592 error-free MCCODs were selected from the 2017 United States multiple cause of death data. The ten most common types of errors in completing the MCCOD (according to published studies) were individually simulated on the error-free MCCODs. After each simulation, the MCCODs were coded using Iris automated mortality coding software. Chance-corrected concordance (CCC) was used to measure the impact of certification errors on the underlying cause of death. Weights for each error type and Socio-demographic Index (SDI) group (representing different mortality conditions) were calculated from the CCC and categorised (very high, high, medium and low) to describe their effect on cause of death accuracy. Findings The only very high impact error type was reporting an ill-defined condition as the underlying cause of death. High impact errors were found to be reporting competing causes in Part 1 [of the death certificate] and illegibility, with medium impact errors being reporting underlying cause in Part 2 [of the death certificate], incorrect or absent time intervals and reporting contributory causes in Part 1, and low impact errors comprising multiple causes per line and incorrect sequence. There was only small difference in error importance between SDI groups. Conclusions Reporting an ill-defined condition as the underlying cause of death can seriously affect the coding outcome, while other certification errors were mitigated through the correct application of mortality coding rules. Training of physicians in not reporting ill-defined conditions on the MCCOD and mortality coders in correct coding practices and using Iris should be important components of national strategies to improve cause of death data quality.

BackgroundThe widely-used estimates of completeness of birth registration collected by Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) and published by UNICEF primarily rely on registration status of children as reported by respondents. However, these self-reported estimates may be inaccurate when compared with completeness as assessed from nationally-reported official birth registration statistics, for several reasons, including over-reporting of registration due to concern about penalties for non-registration. This study assesses the concordance of self-reported birth registration and certification completeness with completeness calculated from civil registration and vital statistics (CRVS) systems data for 57 countries.MethodsSelf-reported estimates of birth registration and certification completeness, at ages less than five years and 12-23 months, were compiled and calculated from the UNICEF birth registration database, DHS and MICS. CRVS birth registration completeness was calculated as birth registrations reported by a national authority divided by estimates of live births published in the United Nations (UN) World Population Prospects or the Global Burden of Disease (GBD) Study. Summary measures of concordance were used to compare completeness estimates.FindingsBirth registration completeness (based on ages less than five years) calculated from self-reported data is higher than that estimated from CRVS data in most of the 57 countries (31 countries according to UN estimated births, average six percentage points (p.p.) higher; 43 countries according to GBD, average eight p.p. higher). For countries with CRVS completeness less than 95%, self-reported completeness was higher in 26 of 28 countries, an average 13 p.p. and median 9-10 p.p. higher. Self-reported completeness is at least 30 p.p. higher than CRVS completeness in three countries. Self-reported birth certification completeness exhibits closer concordance with CRVS completeness. Similar results are found for self-reported completeness at 12-23 months.ConclusionsThese findings suggest that self-reported completeness figures over-estimate completeness when compared with CRVS data, especially at lower levels of completeness, partly due to over-reporting of registration by respondents. Estimates published by UNICEF should be viewed cautiously, especially given their wide usage.

Background and objectiveMany countries have used the new ANACONDA (Analysis of Causes of National Death for Action) tool to assess the quality of their cause of death data (COD), but no cross-country analysis has been done to verify how different or similar patterns of diagnostic errors and data quality are in countries or how they are related to the local cultural or epidemiological environment or to levels of development. Our objective is to measure whether the usability of COD data and the patterns of unusable codes are related to a country's level of socio-economic development.MethodsWe have assessed the quality of 20 national COD datasets from the WHO Mortality Database by assessing their completeness of COD reporting and the extent, pattern and severity of garbage codes, i.e. codes that provide little or no information about the true underlying COD. Garbage codes were classified into four groups based on the severity of the error in the code. The Vital Statistics Performance Index for Quality (VSPI(Q)) was used to measure the overall quality of each country's mortality surveillance system.FindingsThe proportion of 'garbage codes' varied from 7 to 66% across the 20 countries. Countries with a high SDI generally had a lower proportion of high impact (i.e. more severe) garbage codes than countries with low SDI. While the magnitude and pattern of garbage codes differed among countries, the specific codes commonly used did not.ConclusionsThere is an inverse relationship between a country's socio-demographic development and the overall quality of its cause of death data, but with important exceptions. In particular, some low SDI countries have vital statistics systems that are as reliable as more developed countries. However, in low-income countries, where most people die at home, the proportion of unusable codes often exceeds 50%, implying that half of all cause-specific mortality data collected is of little or no use in guiding public policy. Moreover, the cause of death pattern identified from the data is likely to seriously under-represent the true extent of the leading causes of death in the population, with very significant consequences for health priority setting. Garbage codes are prevalent at all ages, contrary to expectations. Further research into effective strategies deployed in these countries to improve data quality can inform efforts elsewhere to improve COD reporting systems.

INTRODUCTION:Adult male and female mortality declines in Japan have been slower than in most high-income countries since the early 1990s. This study compares Japan's recent life expectancy trends with the more favourable trends in Australia, measures the contribution of age groups and causes of death to differences in these trends, and places the findings in the context of the countries' risk factor transitions. METHODS:The study utilises data on deaths by age, sex and cause in Australia and Japan from 1950-2016 from the Global Burden of Disease Study. A decomposition method measures the contributions of various ages and causes to the male and female life expectancy gap and changes over four distinct phases during this period. Mortality differences by cohort are also assessed. FINDINGS:Japan's two-year male life expectancy advantage over Australia in the 1980s closed in the following 20 years. The trend was driven by ages 45-64 and then 65-79 years, and the cohort born in the late 1940s. Over half of Australia's gains were from declines in ischaemic heart disease (IHD) mortality, with lung cancer, chronic respiratory disease and self-harm also contributing substantially. Since 2011 the trend has reversed again, and in 2016 Japan had a slightly higher male life expectancy. The advantage in Japanese female life expectancy widened over the period to 2.3 years in 2016. The 2016 gap was mostly from differential mortality at ages 65 years and over from IHD, chronic respiratory disease and cancers. CONCLUSIONS:The considerable gains in Australian male life expectancy from declining non-communicable disease mortality are attributable to a range of risk factors, including declining smoking prevalence due to strong public health interventions. A recent reversal in life expectancy trends could continue because Japan has greater scope for further falls in smoking and far lower levels of obesity. Japan's substantial female life expectancy advantage however could diminish in future because it is primarily due to lower mortality at old ages.

Many national and subnational governments need to routinely measure the completeness of death registration for monitoring and statistical purposes. Existing methods, such as death distribution and capture-recapture methods, have a number of limitations such as inaccuracy and complexity that prevent widespread application. This paper presents a novel empirical method to estimate completeness of death registration at the national and subnational level.Random-effects models to predict the logit of death registration completeness were developed from 2,451 country-years in 110 countries from 1970-2015 using the Global Burden of Disease 2015 database. Predictors include the registered crude death rate, under-five mortality rate, population age structure and under-five death registration completeness. Models were developed separately for males, females and both sexes.All variables are highly significant and reliably predict completeness of registration across a wide range of registered crude death rates (R-squared 0.85). Mean error is highest at medium levels of observed completeness. The models show quite close agreement between predicted and observed completeness for populations outside the dataset. There is high concordance with the Hybrid death distribution method in Brazilian states. Uncertainty in the under-five mortality rate, assessed using the dataset and in Colombian departmentos, has minimal impact on national level predicted completeness, but a larger effect at the subnational level.The method demonstrates sufficient flexibility to predict a wide range of completeness levels at a given registered crude death rate. The method can be applied utilising data readily available at the subnational level, and can be used to assess completeness of deaths reported from health facilities, censuses and surveys. Its utility is diminished where the adult mortality rate is unusually high for a given under-five mortality rate. The method overcomes the considerable limitations of existing methods and has considerable potential for widespread application by national and subnational governments.